E-MU 0404 USB Modification?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
There is no point in modding this unit unless one can get rid of the 110kHz generated by the switching power supply inside.

I'm not sure if I understand what you are talking about? 110khz? Could you perhaps point to certain components on the PCB? Anyway, I'm planning on using an old ZIP-drive PSU instead of the wall wart that came with the 0404. Don't know if it makes any difference but testing it doesn't cost more than the connector.

Even on spdif output with an exteranal dc supply, the unit is far inferior sounding to a $75Musiland US 01 which is capable of 192k output.

I haven't seen the output on an oscilloscope but to distinguish between two devices I would suspect that maybe the difference has something to do with the device firmware/drivers. I doubt that there's a difference that would remain in ABX testing...

The software on the 0404 also sucks and this is typical of Creative who supplies hundreds of MB of Junk Software which is not necessary.

Sure, the SW package is a bit large but the SW seems solid. I have never had problems with the 1212M and the 0404 seems to be working just fine once I defined the sample rate in Windows correctly (so that the device doesn't try to change the sample rate on the fly... I have a bit of a mixed setup).
 
Last edited:
I'm not sure if I understand what you are talking about? 110khz? Could you perhaps point to certain components on the PCB? Anyway, I'm planning on using an old ZIP-drive PSU instead of the wall wart that came with the 0404. Don't know if it makes any difference but testing it doesn't cost more than the connector.



I haven't seen the output on an oscilloscope but to distinguish between two devices I would suspect that maybe the difference has something to do with the device firmware/drivers. I doubt that there's a difference that would remain in ABX testing...



Sure, the SW package is a bit large but the SW seems solid. I have never had problems with the 1212M and the 0404 seems to be working just fine once I defined the sample rate in Windows correctly (so that the device doesn't try to change the sample rate on the fly... I have a bit of a mixed setup).

Look at the input voltage terminals with a good scope; doesnt matter what you use as dc source, the 110kHz modulation is there.
 
Since I have a Sennheiser HD-650, I am quite interesting in this discussion. Talking about DC-DC converter, it works just like a switching mode power supply: we can convert the voltage up, down or even revert the polarity of the input 5V supply. From what I read so far, it looks like this "SMPS"'s working frequency is too low: 110KHz. This is definitely a good sound killer without any doubt (even it is working on 10MHz freq, you still can hear the difference, trust me)! However, using some ferrite bead, ceramic caps around those DC-DC converter (of course, you need to cut the traces), I think it is still possible to improve 0404 USB to a quite decent headphone amp. Again, how to implement this modification, we need to get a schematics first. If you feel the sound is lean and dry, that is what we are talking about: 110KHz DC-DC converter operating freq. and its higher order harmonics.
 
Schematics for EMU 0404 USB

Hi, I'm interested in extensively modding my 0404 USB, and for that I need its circuit schematics, as detailed as I can get them. If anyone has these, I would appreciate a PM or a link to them.

I am also seeking schematic and block diagram. I am using emu 0404 usb and Tracker for measurements, primary reason is that on 44.1kHz, 24bit, MIN gain settings I get fairly good dynamic range. Tracker gives me better isolation fro 110V line, while its input noise is considerable worse than EMU 0404 USB.

Recently I deal with recording signals in frequency band below 0.1Hz - 0.01Hz. I am desperate enough to get into the endstage circuit in order to avoid all coupling capacitors and bypass low quality circuits between Lin In - ADC. Has anyone injected signal directly into ADC? Improvement noticed?
 

iko

Ex-Moderator
Joined 2008
That's good info, thanks Jim!

For what it's worth, I've seen high frequency ripple on the -5V rail (about 160kHz) with the card connected to absolutely nothing. I'm not sure, but this HF ripple may be a side-effect of the DC-DC inverting converter, the TPS6755, which works on a switching frequency of 160kHz.
 
Jim- good work. My 0202 has the same noisy output and I was a bit confounded trying to reconcile the published specs with what I saw on the scope. I was able to reduce it by changing computer settings (can't remember what) but it was still useless as a signal source. IMO, there's some synchronization going on when using Visual Analyzer and good measurements can be made, but the stuff is still there. Thus my basic distrust of sound card based measurements unless some traditional means can be used to verify at least a couple points. My results have always been very set-up specific.

Best,
Conrad
 
I have performed some mods in my EMU 0404 USB too. I have changed all those crappy electrolytics caps for some nices Panasonics FM and FC. I still have to change the output opamps and the headphone opamps to LM4562. I am a bit afraid of skrewing up something because of the tiny smd components.

Another big improvement will be bypassing the output pot. Seems to be a really cheap one, in RMAA tests it showed a bit unbalance in tracking between channels (depending of pot position).
Add to that the fact that I have been using it as a preamp through a daily whole year usage, that means it should sound significatly worse than when it was factory new.
I think I will bridge the pot and use my brand new DCB1 preamp with stepped attenuator.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Jim- good work. My 0202 has the same noisy output and I was a bit confounded trying to reconcile the published specs with what I saw on the scope. I was able to reduce it by changing computer settings (can't remember what) but it was still useless as a signal source. IMO, there's some synchronization going on when using Visual Analyzer and good measurements can be made, but the stuff is still there. Thus my basic distrust of sound card based measurements unless some traditional means can be used to verify at least a couple points. My results have always been very set-up specific.

Best,
Conrad

I have seen high output noise on my scope too from my Emu tracker. Even sine waves produced by it are obviously dirty VS my analogue old generator on the scope's screen. But when looped with measurement software, it ignores it, and it compares very well for THD readings and SNR of DUT VS normal measurement equipment. A guy in a local forum made an extensive analysis and comparison with his HP gear, and beyond square wave generation that was ringing, he found the EMU very useful when looped with measurement software, but not as a stand alone generator. Especially not beyond 48kHz sampling rate. Big noise starts at 35kHz and he made a filter. That helped if there is need for an AF only capable stand alone generator and it must be the EMU, for economic and utilization reasons if its already in hand. They also proposed buffered attenuation for its input so it can be used on measuring high voltage signals too.
 
Looks very interesting, really nice information. You gave me the idea to do something similar. What of the approach for outputs have you used? You explain it with a transformer, with discrete solid state, hybrid and pure tubes.
Have you constructed and tried all of them? What differences have you noticed.

Leaving still connected the old output stage, is not gonna make some kind of influence? Like input impedance is divided if you place 2 output stages in parallel.

Regards,
Regi
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.